1. Field of the Invention
The invention relates to infrared instruments and cameras and in particular to calibration of the same.
2. Description of the Prior Art
Recent advances in infrared (IR) detector technology have led to the commercially availability of new detector technologies that do not require cryogenic cooling. These new detector technologies include ferroelectric and microbolometer technologies. The currently commercially available microbolometer detector arrays are based on a technology that was originally developed by Honeywell Sensor and System Development Center and was licensed to Boeing (formerly Rockwell), Raytheon (formerly Hughes"" Santa Barbara Research Center) and British Aerospace (formerly Loral).
While these new technologies do not require cryogenic cooling, they are traditionally operated with a thermoelectric cooler to maintain a constant focal plane temperature usually between 0xc2x0 to 20xc2x0 C. The purpose of maintaining a constant detector temperature is that this will simplify the interpretation of the signal generated by the detector array and the generation of a thermal image. Since the detector is at a constant temperature, any measured voltage changes must correspond to thermal differences in the scene that is being imaged onto the focal plane array (focal plane array). The resulting simplification in processing is a result of assuming that the measured temperature differences and the temperature of focal plane array are constant. The apparent temperature from an object in the scene as compared to the average temperature of the scene at the focal plane array are small as compared to the average temperature of the scene (xcex94T  less than  less than T). Assuming the temperature of the focal plane array is constant, then it can be assumed that the measured changes in voltage, that are the result of resistivity changes, are considered to be approximately linear as related to temperature. Since the transform is based on a linear relationship, then the slight differences in the responsivity of the individual elements in the array are typically corrected using a two point calibration that determines an offset and gain correction for each pixel. The results of this two point calibration are independent of the ambient temperature.
While there is a potential of using an uncooled microbolometer focal plane array without a thermoelectric cooler, the difficulty is that calculating the conversion between the measured voltage to a perceived scene temperature difference is now a more complex calculation. The relationship between the voltage and the temperature varies with the temperature of the focal plane array, and the specifics of this variation are different for every pixel in the array. Since the focal plane array temperature is unregulated, just viewing an IR scene will result in heating of the focal plane array.
The invention is directed to a method to vary the gain, offset and/or bias correction tables as a function of ambient temperature without the use of intensive digital numeric processing and to an apparatus for performing the same. The invention corrects for the temperature induced non-uniform response of a array of detectors in a focal plane array (focal plane array) without the use of digital processing and/or computations. The invention permits the operation of a temperature dependent focal plane array without a temperature stabilization cooler and/or heater over a wide range of ambient temperatures. The method of the invention comprises an electronic implementation using high density flash random access memory, RAM, as the memory storage medium for the calibration database, accessing this database based digitized temperature measurements, and then using the bias, gain and offset data within the database to correct for focal plane array response variation induced by temperature changes. The calibration database is comprised of an array of bias, gain and offset values for each pixel in the sensor for each potential operating temperature over the entire range of potential operating temperatures.
One of the innovations incorporated into the invention is based on using limited analog processing and the digitized response of reference temperature measurements from the focal plane array, as an index marker that can be used to directly access the appropriate bank of flash memory that contains the appropriate gain, offset, and bias settings that can either be read directly into the focal plane array via its read-out integrated circuit or can be performed using simple analog or digital circuits.
One benefit of the invention is that imagery generated from an focal plane array, whose response varies non-uniformly with temperature will be consistent even as the ambient temperature varies. The illustrated embodiment explicitly contemplates a long wave infrared (LWIR) microbolometer focal plane array typically operating in the range of 8-14 xcexcm and a short wave infrared (SWIR) InGaAs focal plane array typically operating in the range of 0.7-1.5 xcexcm such as Sensors Unlimited, Inc.""s model number SU320-1.7T1. However, any focal plane array now known or later devised may benefit from application of the present invention. This approach for compensating for ambient temperature variation will significantly reduce the power as compared to the traditional approach based on using a thermoelectric cooler to regulate the temperature of the focal plane array or using a processor to calculate the values in real time.
The methodology of the invention is based on using flash memory to retain all potential bias, gain and offset values. This memory is accessed via digitization of a temperature measurement with no digital processing required. As a result the size, power consumption and production cost of the camera is significantly reduced.
This invention provides a compact, low-power-consumption, commercially viable approach to compensate for temperature induced non-uniform response of the focal plane array. This methodology has the ability to reduce the size, weight and production cost of LWIR cameras based on microbolometer focal plane array technology and SWIR cameras based on InGaAs focal plane array technology, while improving reliability. The use of this temperature compensation approach will result in the elimination of cooling and digital processing electronics without sacrificing overall system performance.
The gain, offset and/or bias for each pixel of the focal plane array is determined at each small temperature increment over the entire operating temperature range. The size of this small temperature increment is based on temperature measurement accuracy and the minimum temperature increment that results in a significant change in the focal plane array""s response. This data is stored in high density flash memory such that:
The pixel order in which the data is stored is the same as the pixel order addressed by the read-out integrated circuit on the focal plane array;
Data for each temperature increment is stored in a separate and discrete addressable page of the flash memory; and
The address of these pages of data in flash memory are stored in temperature increment order.
The resulting flash memory database can then be accessed by:
Measuring the ambient temperature on the focal plane array;
Digitizing this temperature measurement using an analog to digital converter (ADC);
Applying a fixed offset and gain value to this digitized temperature reading that results in calculating the flash memory address; and
The address is then used as a pointer to the portion of memory that should be read out, where the portion of memory that is read out is the data for the measured ambient temperature of the focal plane array.
While the method may be described below for the sake of grammatical fluidity as steps, it is to be expressly understood that the claims are not to be construed as limited in any way by the construction of xe2x80x9cmeansxe2x80x9d or xe2x80x9cstepsxe2x80x9d limitations under 35 USC 112, but to be accorded the full scope of the meaning and equivalents of the definition provided by the claims. The invention can be better visualized by turning now to the following drawings wherein like elements are referenced by like numerals.